TSA Boondoggles: High Costs, Low Effectiveness, But It’s Only Your Money
Milton Friedman often observed that no one spends other people’s money as carefully as he spends his own. You don’t need to be an economics professor to understand that – it’s simple human nature.
Regulatory agencies provide an endless stream of examples to prove Friedman’s point. Let’s take a look at one of America’s least favorite agencies, the Transportation Security Administration.
The TSA has an annual budget of $7.9 billion and employs 62,000 people. Its principal function is to operate the passenger screening function at more than 450 commercial airports. Keeping dangerous people off airplanes is unquestionably important, but is it wise to entrust this to a federal bureaucracy? There is solid evidence that the TSA is not very good at this job, but spends a lot of money uselessly.
Exhibit A is the Screening of Passengers by Observation Techniques (SPOT, naturally) program. The idea behind SPOT is that government observers in airports can detect individuals who are intent on terrorism merely by looking at them and discerning behavioral clues. SPOT was begun in 2007 and employs some 2,800 TSA personnel.
Does it do any good? According to a report issued last November by the Government Accountability Office (GAO), it does not. “The subjectivity of the SPOT behavioral indicators and variation in BDO (Behavior Detection Officer) referral rates raise questions about the continued use of behavior indicators for detecting passengers who might pose a risk to aviation security,” the study concluded.
The GAO concluded that the TSA’s study purporting to validate this approach was badly flawed and recommended defunding SPOT.
So did Judicial Watch, whose president, Tom Fitton was quoted in this Washington Post article as saying, “The TSA continues to treat all fliers as a potential threat. They’re just looking at behavior and not looking into other things, like travel patterns. They’re not focused on what they ought to be focused on.”
Getting a government agency to stop wasting money is like pulling teeth – only much, much harder. Instead of ending SPOT, the TSA actually expanded it in April, adding BWI International, as we read in this Washington Times piece.
In this, like almost everything done by government, officials seem to be utterly incapable of admitting that they’ve made a mistake and correcting it.
If TSA personnel are not able to identify likely terrorists before they reach the final screening phase, the machines that supposedly detect dangerous objects on a person aren’t much better, which brings us to Exhibit B, namely the infamous “nude” scanners.
Starting in 2009, the TSA deployed Rapiscan Secure 1000 full-body scanners at many airports, phasing them out in 2013. The acquisition of the machines, training for workers, and associated personnel costs (each requires a team of five workers) has exceeded $1 billion.
Those machines raised serious concerns about privacy (the images are, by design, pretty revealing) and health (putting people through a dose of X-rays isn’t good, and in addition the machines emit ionizing radiation all around), but TSA justified them by claiming that the problems were de minimis and the machines did an excellent job of detecting potentially harmful items. The TSA claimed that the machines were “the best opportunity to detect metallic and non-metallic anomalies concealed under clothing without the need to touch the passenger.”
In August, a careful academic study was published, demonstrating that the scanners were not nearly as effective as people had been led to believe. Nine researchers representing three universities (UC San Diego, University of Michigan, and Johns Hopkins) bought a Rapiscan Secure 1000 on eBay and then worked extensively with it, trying to determine if resourceful and patient adversaries could find ways of getting weapons and explosives past it.
After extensive testing, the team concluded, “While the device performs well against naïve adversaries, fundamental limitations of backscatter imaging allow more clever attackers to defeat it. We show that an adaptive adversary, with the ability to refine his techniques based on experiment can confidently smuggle contraband past the scanner by carefully arranging it on his body, obscuring it with other materials, or properly shaping it.”
Although the Rapiscan 1000 has been replaced in airports by other scanning machines, they are still in use in other government buildings including courts and prisons.
Nor should the fact that they have been replaced be of much comfort, since, as the authors of the study noted, in 2011 two researchers “reverse engineered” the characteristics of the Rapiscan 1000, from which they concluded that it was “very likely that a large (15-20 cm in diameter), irregularly shaped, cm-thick pancake [of plastic explosive] …taped to the abdomen, would be invisible to this technology.” That is exactly what the nine researchers found from their experiments.
Therefore, it would be a mistake to believe that terrorists cannot figure out how to defeat the technology we now rely on.
One of the lead researchers on the paper, Professor J. Alex Halderman of the University of Michigan told Bloomberg that the federal testing was deficient. When evaluating the machines for their ability to detect explosives, the tests only looked at people carrying blocks of C-4, rather than smaller, shaped portions. “They weren’t actually trying to think like an attacker who would modify their techniques based on how the machine was developed,” Halderman said.
These are just two of the many reasons to lament our hasty embrace of the federalization of almost every aspect of security since the 9/11 disasters.
Chris Edwards of Cato Institute hit the nail on the head when he recently wrote, “It’s a problem that the government has a monopoly on aviation security and that TSA is not accountable to anyone for its level of efficiency or performance. Well, it’s accountable to Congress I suppose, but that doesn’t really amount to much these days.”
Article originally published on Forbes.com.